{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/rotationnet-joint-object-categorization-and","title":"RotationNet: Joint Object Categorization and Pose Estimation Using Multiviews from Unsupervised Viewpoints","arxiv_id":"1603.06208","date":"2016-03-20","proceeding":"CVPR 2018 6","authors":["Asako Kanezaki","Yasuyuki Matsushita","Yoshifumi Nishida"],"abstract":"We propose a Convolutional Neural Network (CNN)-based model \"RotationNet,\"\nwhich takes multi-view images of an object as input and jointly estimates its\npose and object category. Unlike previous approaches that use known viewpoint\nlabels for training, our method treats the viewpoint labels as latent\nvariables, which are learned in an unsupervised manner during the training\nusing an unaligned object dataset. RotationNet is designed to use only a\npartial set of multi-view images for inference, and this property makes it\nuseful in practical scenarios where only partial views are available. Moreover,\nour pose alignment strategy enables one to obtain view-specific feature\nrepresentations shared across classes, which is important to maintain high\naccuracy in both object categorization and pose estimation. Effectiveness of\nRotationNet is demonstrated by its superior performance to the state-of-the-art\nmethods of 3D object classification on 10- and 40-class ModelNet datasets. We\nalso show that RotationNet, even trained without known poses, achieves the\nstate-of-the-art performance on an object pose estimation dataset. The code is\navailable on https://github.com/kanezaki/rotationnet","url_abs":"http://arxiv.org/abs/1603.06208v4","url_pdf":"http://arxiv.org/pdf/1603.06208v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"rotationnet-joint-object-categorization-and","repo_url":"https://github.com/kanezaki/rotationnet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"3d-object-classification","task_name":"3D Object Classification"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-categorization","task_name":"Object Categorization"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}